Russia hacked lightweight FBI radios

[jonwienke]

The Brits obtained one of the initial versions of enigma, but were able to break later versions with just known-plaintext attacks and code key books.

 

[RFI-EMI-GUY]

The Brits obtained one of the initial versions of enigma, but were able to break later versions with just known-plaintext attacks and code key books.

Right. One of those is a statistical attack where you look for letters most frequently occurring and the order of occurrence.

Likewise in P25 there will be certain recurring data from the vocoder such as silence, and fixed embedded signalling that can be used as known plaintext flags.

The point is, attacking the entire permutations of the 256 bit key is not the best expenditure of time and effort.

If the NSA were to ever successfully crack an AES256 implementation, or knew that an adversary had, they would not make that information public. But you can be sure somewhere, someone is working on it.

 

[jonwienke]

In the case of the later break, the Germans added additional rotors to the machines, but because they got sloppy and didn't add random padding to the beginning of the message, the Brits used the captured code books and known message headers to reverse engineer the wiring of the extra rotors without ever having one in their possession. An excellent read on the subject:

Colossus: The secrets of Bletchley Park's code-breaking computers Amazon.com: Colossus: The secrets of Bletchley Park's code-breaking computers (8601300147444): B. Jack Copeland: Books

 

[slicerwizard]

Likewise in P25 there will be certain recurring data from the vocoder such as silence, and fixed embedded signalling that can be used as known plaintext flags.

Which doesn't do you much good when every voice superframe uses a different random 64 bit MI.

The point is, attacking the entire permutations of the 256 bit key is not the best expenditure of time and effort.

That's about all you're going to do with your known plaintext...

 

[blantonl]

I think everyone got off track here in this discussion.

There is absolutely no evidence whatsoever that the actual communications by the agents working their details was compromised.

Rather, the Russians were able to use proximity RF management, heuristics, and P25 metadata to determine whether or not FBI counterintelligence teams were active for specific individuals.

 

[slicerwizard]

^^^ What he said. ^^^

I'd like to add that using the cracking of a brain dead encryption algorithm like Enigma as in any way suggesting that AES will also be cracked - is, to be frank, ridiculous. I'd also note that as far as I'm aware, to date, some algorithms with less pedigree, like DES and RC4, have not even been cracked. Note that brute forcing is not cracking. No one is claiming they can break 256 bit RC4, are they?

 

[RFI-EMI-GUY]

^^^ What he said. ^^^

I'd like to add that using the cracking of a brain dead encryption algorithm like Enigma as in any way suggesting that AES will also be cracked - is, to be frank, ridiculous. I'd also note that as far as I'm aware, to date, some algorithms with less pedigree, like DES and RC4, have not even been cracked. Note that brute forcing is not cracking. No one is claiming they can break 256 bit RC4, are they?

Never say never.

To be clear DES was brute force "cracked" 21 years after it was announced. Advances in computational power enabled the brute force

EFF DES cracker - Wikipedia

en.wikipedia.org

Where will computational power be in another 20 years?

You don't necessarily have to go through every permutation of the key space. You can get "lucky".

Moore's law - Wikipedia

en.wikipedia.org

 

[jonwienke]

DES has a 56 bit key. AES has a 256-bit key. Moore's law is not going to continue unchecked for another 300 years, which is what would be required to brute-force the extra 200 bits of keyspace.

 

[slicerwizard]

Where will computational power be in another 20 years?

It will be nowhere near even making a dent in the AES256 keyspace. Moore's Law doesn't help you when brute forcing AES256 requires more energy than is contained in the entire Universe.

You don't necessarily have to go through every permutation of the key space. You can get "lucky".

You're not beating the trillion-trillion-trillion... to one odds unless the key maker deliberately chose a weak key (like 000....0001)

And such a lucky guess is not cracking.

 

[KevinC]

You're not beating the trillion-trillion-trillion... to one odds unless the key maker deliberately chose a weak key (like 000....0001)

So I guess I should change mine...

 

[slicerwizard]

Too late! Cracked open like a walnut...

 

[prcguy]

Wow, all the energy that is contained in the entire universe right in my hand. Marvel makes movies about stuff like that. I could be...... AES 256 Man, Super, Super Hero! My keyloader will take over the entire universe!!!!

It will be nowhere near even making a dent in the AES256 keyspace. Moore's Law doesn't help you when brute forcing AES256 requires more energy than is contained in the entire Universe.


You're not beating the trillion-trillion-trillion... to one odds unless the key maker deliberately chose a weak key (like 000....0001)

And such a lucky guess is not cracking.

 

[prc117f]

Quantum computing is where it will be.

 

[jonwienke]

At this point, nobody expects quantum computers to crack AES.

 

[INDY72]

Just to end this, only the cheapest folks will even have a traditional HT/Mobile/Base radio by the time true Quantum Computers roll out for use. Everything will be done on Cellular Based LTE supersytems. Is that not what you all proclaim daily? FirstNET and future iterations are the end of regular radio? So this is really just a academic historical thing for study in college IT and Telecommunications courses. Much like we all studied Marconi etc? "Yes! I just cracked AES256 on a Motorola ASTRO-25 TRS!" "Oh? how quaint. Actual radio systems and P25, and AES256 were so 2019....." All discussed using implanted chipset com systems in a local Community College course...

 

[RFI-EMI-GUY]

It will be nowhere near even making a dent in the AES256 keyspace. Moore's Law doesn't help you when brute forcing AES256 requires more energy than is contained in the entire Universe.


You're not beating the trillion-trillion-trillion... to one odds unless the key maker deliberately chose a weak key (like 000....0001)

And such a lucky guess is not cracking.

Dont discount that possibility, stupid Humans at work once again;

Equifax used 'admin' as username and password for sensitive data: lawsuit

A class action lawsuit filed in January 2019 claims Equifax used "admin" as both password and username for a portal with sensitive information.

finance.yahoo.com

 

[jonwienke]

There will always be idiots.

 

[slicerwizard]

I don't discount the possibility of peeps choosing weak keys - in fact I've seen it from time to time. It's still not cracking. And it's your only hope for AES.

 

[batdude]

of course, a roll of $100 bills in exchange for access to the KVL... or even a valid radio on the system... is a hell of a lot easier than trying to bust the algo key.

 

[citiot]

^^^^^ yup, meatware (humans) are a lot easier to exploit than hardware and software....